An ensemble machine learning approach for prediction and optimization of modulus of elasticity of recycled aggregate concrete

Han, Taihao; Siddique, Ashfia; Khayat, Kamal H.; Huang, Jie; Kumar, Aditya

doi:10.1016/j.conbuildmat.2020.118271

Cited by 159 publications

(69 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This could be explained by its being an ensemble algorithm. The findings are in corroboration with Han et al [9] who assert that ensemble algorithms tend to perform better than stand-alone algorithms. However, GBM and XGB gave lower accuracy than the DT classifier unlike expectation.…”

Section: Resultssupporting

confidence: 88%

A Comparative Analysis of Machine Learning Models for Prediction of Insurance Uptake in Kenya

Yego¹,

Kasozi²,

Nkrunziza³

2020

Preprint

View full text Add to dashboard Cite

The role of insurance in financial inclusion as well as in economic growth is immense. However, low uptake seems to impede the growth of the sector hence the need for a model that robustly predicts uptake of insurance among potential clients. In this research, we compared the performances of eight (8) machine learning models in predicting the uptake of insurance. The classifiers considered were Logistic Regression, Gaussian Naive Bayes, Support Vector Machines, K Nearest Neighbors, Decision Tree, Random Forest, Gradient Boosting Machines and Extreme Gradient boosting. The data used in the classification was from the 2016 Kenya FinAccess Household Survey. Comparison of performance was done for both upsampled and downsampled data due to data imbalance. For upsampled data, Random Forest classifier showed highest accuracy and precision compared to other classifiers but for down sampled data, gradient boosting was optimal. It is noteworthy that for both upsampled and downsampled data, tree-based classifiers were more robust than others in insurance uptake prediction. However, in spite of hyper-parameter optimization, the area under receiver operating characteristic curve remained highest for Random Forest as compared to other tree-based models. Also, the confusion matrix for Random Forest showed least false positives, and highest true positives hence could be construed as the most robust model for predicting the insurance uptake. Finally, the most important feature in predicting uptake was having a bank product hence bancassurance could be said to be a plausible channel of distribution of insurance products.

show abstract

Section: Resultssupporting

confidence: 88%

A Comparative Analysis of Machine Learning Models for Prediction of Insurance Uptake in Kenya

Yego¹,

Kasozi²,

Nkrunziza³

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…As a distinct indicator for evaluating the performance of a machine learning model, R (Pearson’s correlation coefficient) or R 2 (coefficient of determination) are useful. Thus, many researchers (e.g., Qi et al (2018) [ 44 ]; Han et al (2020) [ 45 ]; Kannangara et al (2018) [ 27 ]; Kumar et al (2018) [ 21 ]) used the R or R 2 value between the observed and predicted value for assessing the performance of prediction models. In this study, for performance evaluation of the prediction model, the R (correlation coefficient) value, which is given by Equations (3) and (4), between the observed and predicted values was used as the performance indicator.…”

Section: Development Of a Dw Prediction Modelmentioning

confidence: 99%

Development of a Prediction Model for Demolition Waste Generation Using a Random Forest Algorithm Based on Small DataSets

Cha

Moon

Kim

et al. 2020

IJERPH

View full text Add to dashboard Cite

Recently, artificial intelligence (AI) technologies have been employed to predict construction and demolition (C&D) waste generation. However, most studies have used machine learning models with continuous data input variables, applying algorithms, such as artificial neural networks, adaptive neuro-fuzzy inference systems, support vector machines, linear regression analysis, decision trees, and genetic algorithms. Therefore, machine learning algorithms may not perform as well when applied to categorical data. This article uses machine learning algorithms to predict C&D waste generation from a dataset, as a way to improve the accuracy of waste management in C&D facilities. These datasets include categorical (e.g., region, building structure, building use, wall material, and roofing material), and continuous data (particularly, gloss floor area), and a random forest (RF) algorithm was used. Results indicate that RF is an adequate machine learning algorithm for a small dataset consisting of categorical data, and even with a small dataset, an adequate prediction model can be developed. Despite the small dataset, the predictive performance according to the demolition waste (DW) type was R (Pearson’s correlation coefficient) = 0.691–0.871, R2 (coefficient of determination) = 0.554–0.800, showing stable prediction performance. High prediction performance was observed using three (for mortar), five (for other DW types), or six (for concrete) input variables. This study is significant because the proposed RF model can predict DW generation using a small amount of data. Additionally, it demonstrates the possibility of applying AI to multi-purpose DW management.

show abstract

“…The proficiency of the RF model’s in terms of promptly producing high-fidelity predictions of cement hydration kinetics (in the forms of time-dependent cumulative heat release and heat flow rate) is anticipated, given the large body of published research 54 – 56 , 58 , 59 , 73 , 77 , 91 ; in which it has consistently been shown that the RF model—and other decision trees-based models—generally outperform other models that are based on nonlinear regression analyses (e.g., elastic net regression), or an assortment of logistic transfer functions (e.g., artificial neural networks), or data clustering/mapping mechanisms (e.g., support vector machines). To benchmark the prediction performance of the RF model, two additional ML models developed and used in our previous studies 55 , 57 – 59 , 92 —the multilayer perceptron artificial neural network (MLP-ANN) model; and support vector machine (SVM)—were used to produce predictions of heat release behavior, and compared against those produced by the RF model (see Fig.…”

Section: Resultsmentioning

confidence: 99%

“…The random forests (RF) model—based on an ensemble of 100–1000 s of uncorrelated classification-and-regression-trees (CART) 55 , 58 , 59 , 73 —is used in this study. The RF model is premised on: (1) Building uncorrelated decision trees (i.e., CARTs) at a large scale; (2) Grouping CARTs into committees, so as to produce independent outputs; and (3) Averaging outputs produced by all CARTs to estimate the final output 74 .…”

Section: Overview Of the Random Forests Modelmentioning

confidence: 99%

“…Manual adjustment of these hyperparameters could lead to poor prediction performance. Therefore, in this study, the grid-search method 76 —coupled with the tenfold cross-validation method (see “ Results and discussion ” section) 58 , 59 , 77 , 78 —was used to autonomously adjust-and-optimize the two hyperparameters.…”

Section: Overview Of the Random Forests Modelmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning enables prompt prediction of hydration kinetics of multicomponent cementitious systems

Lapeyre

Han

Wiles

et al. 2021

Sci Rep

Self Cite

View full text Add to dashboard Cite

Carbonaceous (e.g., limestone) and aluminosilicate (e.g., calcined clay) mineral additives are routinely used to partially replace ordinary portland cement in concrete to alleviate its energy impact and carbon footprint. These mineral additives—depending on their physicochemical characteristics—alter the hydration behavior of cement; which, in turn, affects the evolution of microstructure of concrete, as well as the development of its properties (e.g., compressive strength). Numerical, reaction-kinetics models—e.g., phase boundary nucleation-and-growth models; which are based partly on theoretically-derived kinetic mechanisms, and partly on assumptions—are unable to produce a priori prediction of hydration kinetics of cement; especially in multicomponent systems, wherein chemical interactions among cement, water, and mineral additives occur concurrently. This paper introduces a machine learning-based methodology to enable prompt and high-fidelity prediction of time-dependent hydration kinetics of cement, both in plain and multicomponent (e.g., binary; and ternary) systems, using the system’s physicochemical characteristics as inputs. Based on a database comprising hydration kinetics profiles of 235 unique systems—encompassing 7 synthetic cements and three mineral additives with disparate physicochemical attributes—a random forests (RF) model was rigorously trained to establish the underlying composition-reactivity correlations. This training was subsequently leveraged by the RF model: to predict time-dependent hydration kinetics of cement in new, multicomponent systems; and to formulate optimal mixture designs that satisfy user-imposed kinetics criteria.

show abstract

An ensemble machine learning approach for prediction and optimization of modulus of elasticity of recycled aggregate concrete

Cited by 159 publications

References 64 publications

A Comparative Analysis of Machine Learning Models for Prediction of Insurance Uptake in Kenya

A Comparative Analysis of Machine Learning Models for Prediction of Insurance Uptake in Kenya

Development of a Prediction Model for Demolition Waste Generation Using a Random Forest Algorithm Based on Small DataSets

Machine learning enables prompt prediction of hydration kinetics of multicomponent cementitious systems

Contact Info

Product

Resources

About